Related papers: Plug-and-Play Training Framework for Preference Op…
Despite significant advances in long-context reasoning by large language models (LLMs), primarily through Online Reinforcement Learning (RL) methods, these approaches incur substantial computational costs and complexity. In contrast,…
The ability of LLMs to represent diverse perspectives is critical as they increasingly impact society. However, recent studies reveal that alignment algorithms such as RLHF and DPO significantly reduce the diversity of LLM outputs. Not only…
While large-scale unsupervised language models (LMs) learn broad world knowledge and some reasoning skills, achieving precise control of their behavior is difficult due to the completely unsupervised nature of their training. Existing…
Reinforcement learning with verifiable rewards has shown notable effectiveness in enhancing large language models (LLMs) reasoning performance, especially in mathematics tasks. However, such improvements often come with reduced outcome…
Multi-objective preference alignment of large language models (LLMs) is critical for developing AI systems that are more configurable, personalizable, helpful, and safe. However, optimizing model outputs to satisfy diverse objectives with…
Direct Preference Optimisation (DPO) is effective at significantly improving the performance of large language models (LLMs) on downstream tasks such as reasoning, summarisation, and alignment. Using pairs of preferred and dispreferred…
Mathematical reasoning presents a significant challenge for Large Language Models (LLMs) due to the extensive and precise chain of reasoning required for accuracy. Ensuring the correctness of each reasoning step is critical. To address…
Recently, Large Language Models (LLMs) have rapidly evolved, approaching Artificial General Intelligence (AGI) while benefiting from large-scale reinforcement learning to enhance Human Alignment (HA) and Reasoning. Recent reward-based…
Iterative preference optimization methods have recently been shown to perform well for general instruction tuning tasks, but typically make little improvement on reasoning tasks (Yuan et al., 2024, Chen et al., 2024). In this work we…
Most Video Large Language Models (Video-LLMs) adopt preference alignment techniques, e.g., DPO~\citep{rafailov2024dpo}, to optimize the reward margin between a winning response ($y_w$) and a losing response ($y_l$). However, the likelihood…
Direct Preference Optimization (DPO) has recently emerged as a simple and effective alternative to reinforcement learning from human feedback (RLHF) for aligning large language models (LLMs) with user preferences. However, existing DPO…
Direct alignment methods are increasingly used for aligning large language models (LLMs) with human preferences. However, these methods suffer from the issues of verbosity and likelihood displacement, which can be driven by the noisy…
Large Vision-Language Models (LVLMs) hold immense potential for complex multimodal instruction following, yet their development is often hindered by the high cost and inconsistency of human annotation required for effective fine-tuning and…
Policy exploration is critical in reinforcement learning (RL), where existing approaches include greedy, Gaussian process, etc. However, these approaches utilize preset stochastic processes and are indiscriminately applied in all kinds of…
Multimodal Large Language Models (MLLMs) excel in generating responses based on visual inputs. However, they often suffer from a bias towards generating responses similar to their pretraining corpus, overshadowing the importance of visual…
Offline preference optimization is a key method for enhancing and controlling the quality of Large Language Model (LLM) outputs. Typically, preference optimization is approached as an offline supervised learning task using manually-crafted…
We introduce Direct Value Optimization (DVO), an innovative reinforcement learning framework for enhancing large language models in complex reasoning tasks. Unlike traditional methods relying on preference labels, DVO utilizes value signals…
Direct Preference Optimization (DPO) have emerged as a popular method for aligning Large Language Models (LLMs) with human preferences. While DPO effectively preserves the relative ordering between chosen and rejected responses through…
Traditional language model alignment methods, such as Direct Preference Optimization (DPO), are limited by their dependence on static, pre-collected paired preference data, which hampers their adaptability and practical applicability. To…
DPO (Direct Preference Optimization) has become a widely used offline preference optimization algorithm due to its simplicity and training stability. However, DPO is prone to overfitting and collapse. To address these challenges, we propose…